A ReplicaSet helps load balance and scale our Application up or down when the demand for it changes. It makes sure the desired number of pods are always running for high availability.

I’m using VS Code on Mac to create the below yaml file.

Create the following yaml file:

apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: myapp-replicaset
  labels:
    app: myapp
spec:
  selector:
    matchLabels:
      app: myapp
  replicas: 3
  template:
    metadata:
      name: nginx-2
      labels:
        app: myapp
    spec:
      containers:
        - name: nginx
          image: nginx

The above yaml file has 4 main properties which are apiVersion, kind, metadata and spec. The spec contains the definition for the number of replicas and the containers to be created inside the pods.

Run the following command:

$ kubectl create -f ReplicaSetsDemo.yaml 

replicaset.apps/myapp-replicaset created

Now let’s check the status of the replicaSet and then the pods:

$ kubectl get replicaset
NAME               DESIRED   CURRENT   READY   AGE
myapp-replicaset   3         3         3       2m29s


$ kubectl get pods

NAME                     READY   STATUS    RESTARTS   AGE
myapp-replicaset-s7jm7   1/1     Running   0          33s
myapp-replicaset-svqvm   1/1     Running   0          33s
myapp-replicaset-xnbbq   1/1     Running   0          33s

The above command shows 3 pods created for nginx with the name of the replicaset prefixed. A replicaset ensures that sufficient number of replicas or pods are available at all times.

Now, let’s delete a pod:

$ kubectl delete pod myapp-replicaset-s7jm7

pod "myapp-replicaset-s7jm7" deleted

Check the status of the pods again:

$ kubectl get pods
                         
NAME                     READY   STATUS    RESTARTS   AGE
myapp-replicaset-d7f88   1/1     Running   0          36s
myapp-replicaset-svqvm   1/1     Running   0          5m53s
myapp-replicaset-xnbbq   1/1     Running   0          5m53s

Notice that there are still 3 pods running as one more pod was created to maintain the desired state. Also, if you try to create a pod with the same label app=myapp outside of the replicaset, it’ll still come under the purview of the replicaset we created and will terminate that pod to maintain the desired state of 3 replicas. This is where it differs from a Replication Controller.

Now, let’s edit the replicaset to scale it up as below:

$ kubectl edit replicaset myapp-replicaset

The above command let’s you edit the in-memory configuration file that Kubernetes creates. Search for the replicas section and edit it to 4 using the vi editor commands and save it with wq!

Upon saving you’ll see the output as:

replicaset.apps/myapp-replicaset edited

Check the pods status and you’ll see and additional pod created:

$ kubectl get pods
                            
NAME                     READY   STATUS    RESTARTS   AGE
myapp-replicaset-d7f88   1/1     Running   0          14m
myapp-replicaset-pnlvh   1/1     Running   0          55s
myapp-replicaset-svqvm   1/1     Running   0          19m
myapp-replicaset-xnbbq   1/1     Running   0          19m

Another way to scale the replicaset is to use the below command and check the status again:

$ kubectl scale replicaset myapp-replicaset --replicas=2

replicaset.apps/myapp-replicaset scaled


$ kubectl get pods

NAME                     READY   STATUS    RESTARTS   AGE
myapp-replicaset-svqvm   1/1     Running   0          22m
myapp-replicaset-xnbbq   1/1     Running   0          22m

Notice that the pods are scaled down to 2 only as other 2 got terminated.